#207 · Primary category: Computer Vision

multi-object-tracker

caffe computer-vision deep-learning iou-tracker multi-object-tracking neural-network object-detection object-tracking opencv python python3 ssd-mobilenet tensorflow tracker yolo yolov3

Multi-object trackers in Python

Project last updated:07/17/26

GitHub Stars

696

Forks

101

Contributors

6

License

MIT

Why we included this project

Multi-object tracking is fiddly: you need to match detections across frames without losing identities, and the classic algorithms each have their own quirks. This library wraps CentroidTracker, IOUTracker, CentroidKF_Tracker, and SORT behind one simple API, so you can test different strategies without rewriting your pipeline. It also includes OpenCV dnn-based detectors like YOLOv3 and SSD MobileNet that feed the trackers, which makes it a handy starting point for a full detect-and-track loop. The examples run on real video, and the code is compact enough to read and modify, so it works well both for learning how trackers behave and for getting a working prototype up quickly.

Articles for this project

No articles for this project yet.

To suggest a topic or contribute an article, contact us.

Related projects in this category